NREC built upon its expertise in vision, mechanisms and manipulation to build an automated strawberry plant sorter that streamlines the harvesting process, improves efficiency, and ensures consistent plant quality. Lassen Canyon Nursery and other growers, representing approximately 85% of California’s strawberry plant nursery market, supported this project and plan to use this technology in their operations.
Harvesting strawberry plants is a complex task that’s only partially automated. Workers first use farming equipment to dig mature plants from the ground and remove excess soil from their roots. Farm workers then manually sort and package the plants in trimming sheds. The workers trim, count and pack the high-quality plants for shipment and discard the low-quality ones – all by hand.
Sorting quality strawberry plants from plants that are damaged, of inappropriate size, or otherwise unsuitable for growers is the most costly part of the entire production cycle. To meet growers’ strict quality standards, plants must be of a uniform size. Using an automated sorter will ensure more consistent plant quality and make the harvesting process more efficient.
The vision system inspects the harvested strawberry plant rootstock. It uses machine learning techniques to determine which strawberry plants are shippable and which ones are not. The vision system is trained to sort strawberry plants using samples of plants harvested by a human. The training set teaches it how to classify plants into different categories corresponding to the size and health of the plants.
The system sorts plants of different varieties and levels of maturity and operates under realistic conditions, where rain and frost change plants’ appearance and roots may contain mud and debris.
The system uses air jets to sort the plants into three bins based on their classification. High-quality plants can go on to be packaged and shipped to customers. Plants that do not meet the quality standards can be discarded. Unknown and borderline plants can be sent to a human inspector for manual classification and sorting.
National Robotics Engineering Center
10 40th Street
Pittsburgh, PA 15201
+1 (412) 681-6900
Carnegie Mellon University
Legal Info | www.cmu.edu